I don't really like the idea of just swapping parts without understanding their influence on the problem. Of course it's especially unsatisfying when you can't even notice a difference after doing it.
I decided to do some reading up on the working of the sound board. Altan gives useful insight about the channels in his posts and on his website (https://www.aaarpinball.com/TwilightZone/TwilightZone.htm).
So I quickly found out there are 4 channels on the board:
CVSD (voice samples)
CH1, CH2 (both music)
DAC (not really sure at first, but I found out later this at least is responsible for percussion)
I read somewhere the music on the CH1 and CH2 is generated by two Yamaha chips (YM2151 and YM3012). At first I was kind of annoyed with myself for not sooner swapping out these socketed ICs between boards, because this would be an easy check.
After swapping them out however, there was no audible change in audio quality.
I decided to dig deeper and wanted to see if I could analyse the signals on the channels separately. I attached my oscilloscope and probed:
- U7 at pin 7 for CVSD
- U7 at pin 8 for DAC
- U8 at pin 8 for CH1
- U8 at pin 14 for CH2
With the oscilloscope I could confirm that CVSD was responsible for speech and CH1 and CH2 for music. It also showed the DAC is involved in music playback as well.
I managed to export some signals from my scope to my computer in csv-format and imported them in the free software Audacity.
It made it possible for me to play back short signals and compare two simultaneously sampled channels. This was pretty awesome!
I found out CH1 and CH2 actually sound equal and excellent. I really like the sound of these Yamaha chips!
In the screenshot you can see both channels are indistinguishable. I wonder why two channels are used. Does anybody have an idea?
After comparing the signals of CH1 and DAC I found some erratic signals on the DAC channel, that I show in the following clip.
I made the following observations:
- CH1 is good
- The DAC signal is very asymmetrical, which looks strange to me. Is this due to the place where I tapped the signal or is it part of the problem?
- The DAC signal seems to collapse at certain moments that correspond with audible cracks.
After this I decided to replace some more capacitors (47uf: C32, C35). C35 seems to be closely linked to the DAC circuit.
But to no resolve.
At the moment I am convinced the problem is in or around the DAC (AD7524). I am not ready to replace the chip yet, I still have some questions:
- Is the AD7524 blown or something connected to it?
- Is there a relationship with the burnt-out C47?
- I would like to look at the DAC signal in my good board. Is this signal symmetrical?
Any feedback is highly appreciated!